Can Feedback based on Predictive Data Improve Learners' Passing Rates in MOOCs? A Preliminary Analysis

Abstract

This work in progress paper investigates if timely feedback increases learners' passing rate in a MOOC. An experiment conducted with 2,421 learners in the Coursera platform tests if weekly messages sent to groups of learners with the same probability of dropping out the course can improve retention. These messages can contain information about: (1) the average time spent in the course, or (2) the average time per learning session, or (3) the exercises performed, or (4) the video-lectures completed. Preliminary results show that the completion rate increased 12% with the intervention compared with data from 1,445 learners that participated in the same course in a previous session without the intervention. We discuss the limitations of these preliminary results and the future research derived from them.

Response of learning analytics to the online education challenges during pandemic: Opportunities and key examples in higher education

“…Some studies reported that LA tools have benefits for improving students' retention as a response to increased dropping-out risk ( Perez-Sanagustin et al, 2021 ). Instructors generally have time limitation for detecting students at-risk and offering adaptive feedback for students ( Günther, 2021 ).…”

“…Students at-risk who cannot receive adaptive support are more likely to drop out the course ( Martin et al, 2020 ). In such times, LA tools with the advantages of timely support and intervention can increase student retention ( Perez-Sanagustin et al, 2021 ). It has been also reported that LA tools were perceived as easy to use ( Maher et al, 2020 ; Liu et al, 2020 ).…”

Meta Transfer Learning for Early Success Prediction in MOOCs

“…Given the low success rates for many MOOCs, targeted interventions have the potential to improve learning outcomes [6,[28][29][30]. We are therefore interested in early success prediction, providing the basis for such targeted intervention (e.g., offering additional support to students at risk of failing the course).…”

“…The features representing the Title, the Short Description, and the Long Description were obtained by translating the corresponding text to English via the DeepL translation service and then finetuning a pre-trained FastText model [34] on that specific meta information 6 . FastText is a state-of-the-art model for word representation (i.e., representing words in a text as latent vectors that can be used for machine learning classification tasks).…”

“…Adaptive guidance can hence increase retention and student performance. Methods for early success prediction build the basis for interventions [6].…”